| Age | Commit message (Collapse) | Author |
|---|
|
ok kettenis@ cloder@ tom@ henning@
|
|
Rename pentium_mhz to cpuspeed which is consistant with amd64 making
shared ACPI code less nasty.
ok marco, deraadt
|
|
the remainder of the network stack from splimp to splnet.
ok miod@
|
|
in the data part for the data from the previously distinct tags.
look up the tag early and carry a pointer to it around.
makes the code easier and saves some tag lookups and thus helps performance,
as proven by tests run by Schberle Dniel <Schoeberle.Daniel@aamtech.hu>
Initially hacked up somewhere over the atlantic ocean in an A330
early testing reyk and moritz, "put it in" theo
|
|
code is all conditionalized on __HAVE_TIMECOUNTER, and not
enabled on any platforms.
adjtime(2) support exists, courtesy of nordin@, sysctl(2) support
and a concept of quality for each time source attached exists.
High quality time sources exists for PIIX4 ACPI timer as well as
some AMD power management chips. This will have to be redone
once we actually add ACPI support (at that time we need to use
the ACPI interfaces to get at these clocks).
ok art@ ken@ miod@ jmc@ and many more
|
|
encapsulating all such access into wall-defined functions
that makes sure locking is done as needed.
It also cleans up some uses of wall time vs. uptime some
places, but there is sure to be more of these needed as
well, particularily in MD code. Also, many current calls
to microtime() should probably be changed to getmicrotime(),
or to the {,get}microuptime() versions.
ok art@ deraadt@ aaron@ matthieu@ beck@ sturm@ millert@ others
"Oh, that is not your problem!" from miod@
|
|
reload rules.
this fixes an altq problem that, if you reload pf rules not containing
queues while running altq, the interface shaper is not properly removed.
make pf_altq_running local to pf_ioctl.c since it is no longer used in
altq_subr.c.
ok henning@
|
|
is full, along with a timer that unsets it again after 10ms.
The input queue beeing full is a reliable indicator for CPU overload, and
this flag allows other subsystems to cope with the situation.
hacked with beck
ok kjc@ markus@ beck@
|
|
identifiers without embedded meanings.
this also allows us to make the semantics of the qid assignment in line
with the tag assignment in the next step.
ok, henning@
|
|
the change removes the advertising clause, which was requested by a linux
developer.
ok deraadt@
|
|
|
|
ok dhartmei@ jmc@
|
|
|
|
|
|
hif->hif_rootclass should be initialized when the root queue is created.
|
|
|
|
|
|
control over the root class from pfctl.
also fix a few null pointer derefs in case a root class does not exist
|
|
what we do for teh other schedulers already
|
|
|
|
diff changes that.
with PRIQ, the queues are in an array, with the priority as key. removing
the tie means we cannot access the array with (queueID - 1) as key any
more but need to traverse the array until the queue ID matches. As the array
has a maximum of 16 entries, traversing linear is okay.
a new queue ID allocation algorithm coming soon will require this.
ok dhartmei@ kjc@
|
|
|
|
|
|
ok kjc@
|
|
IDs in userland already.
|
|
Patrick Latifi, Thanks!
|
|
from Patrick Latifi
|
|
in error handling of hfsc_add_altq().
report and fix by millert@
|
|
the min and max thresholds are set to 10% and 30% of the queue limit.
|
|
from KAME
|
|
this should not happen but just in case.
printf() is intended to be annoying so that we'll get reports on it.
original idea from dhartmei@
ok deraadt@, henning@
|
|
validity and take that.
validity checks provided by kjc@
ok dhartmei@ kjc@
|
|
|
|
|
|
"accidently" typed while hunting a bug ;-)
ok kjc@
|
|
ok henning@, deraadt@
|
|
|
|
ok theo
|
|
ok kjc@
|
|
- remove files which are no longer used, or we don't have plans to support
in pf in the near future.
- remove altq ioctl related stuff.
- convert the PRIQ, HFSC and RIO modules to pf-based altq.
(these are not enabled in GENERIC, CDNR is not converted yet.)
|
|
|
|
avoid kernel-rebuild with ALTQ_NOPCC for non-pentium users.
|
|
after altq gets flushed, altq forgot that it was enabled since
altq is actually detached with an empty ruleset.
so, add a variable, pfaltq_running, to remember the running state
and re-enable altq when a new ruleset is loaded.
noticed, tested, and oked by henning@
|
|
|
|
for now, implemnet a tentative transition mechanism to allow the old API
and the new API to coexit.
the old API will go away when pfctl(8) becomes ready.
|
|
|
|
this commit is to allow further development in both userland and kernel.
the goal is to replace altq's classifier by pf(4).
- make pf tag a queue id to mbuf and make altq read the queue id
- merge altq config into pf.conf(5)
ok dhartmei@, henning@
|
|
this bug is triggered only when 3 or more cbq instances are
created at a time.
|
|
|
|
|
